CBS switches to new method for measuring business confidence

© Sjoerd van der Hucht Fotografie
From July 2023, CBS will use a new method to measure business confidence based on the results of the Netherlands Business Survey (COEN). The old method focused on the correlation between business confidence and GDP, which meant that the questions used varied by branch of industry. Under the new method, the same questions are used for all industries, making it easier to draw comparisons. In this article, we discuss all the ins and outs of the switch.

The COEN, a monthly survey commissioned by the European Commission, maps the mood of Dutch entrepreneurs by asking them questions on how they feel things went in the previous three months and how they expect the next three months to go. In the first month of each quarter, a number of additional industries are surveyed and the survey itself is more comprehensive, providing information in greater detail. Statistics Netherlands (CBS) conducts this quarterly survey in cooperation with the Chamber of Commerce, the Economic Institute for Construction and Housing (EIB), the Dutch Employers’ Organisation (VNO-NCW) and the Dutch Organisation for Small and Medium-Sized Enterprises (MKB-Nederland).

Business confidence is measured quarterly using the results of the COEN, based on a number of balances. These balances are calculated by subtracting the percentage of entrepreneurs who report negative experiences or expectations from the percentage that report positive experiences or expectations. The balances are then adjusted for seasonal effects and averaged without weighting to create business confidence indices for each branch of industry.

Old method: correlation with GDP but limited comparability

The old method focused on the correlation between business confidence and GDP. Partly for this reason, the questions used to calculate business confidence varied across industries. When the old method was developed, the questions used for each branch of industry were selected to ensure correlation with the development of GVA. This meant that different questions were used for manufacturing than for construction, and that there was a gap between the industrial sector as a whole and the service industries. In addition, the business confidence balances for some industries were corrected to bring the average of the series to zero, while this was not done for other branches. As a result, the business confidence indices for different branches of industry were only comparable to a limited extent, which in turn meant that the business confidence index for the non-financial private sector as a whole was difficult to interpret.

New method: consistent and comparable sentiment indicator

Consistency and comparability are important when it comes to the usability of business confidence figures. Under the new method, the same two questions – which are thematically aligned with the updated definition of the sentiment indicator – are used across all branches of industry. As with the old method, a balance is calculated for each question, but the questions are no longer based on their correlation to GDP. The business confidence index now mainly serves as a sentiment indicator for the Dutch non-financial private sector. While there is still a correlation between the historical series and GDP, this is less defined than under the old method.

The two chosen questions pertain to the development of the economic environment in the past three months and the respondent’s expectations for the next three months, and the balances are adjusted only for seasonal effects. This makes business confidence figures for different branches of industry highly comparable and easy to break down by region and firm size. Based on the branch figures, a consistent figure for the non-financial private sector as a whole can be calculated using weighted averages based on the number of persons employed.

The new method will not only make business confidence figures more consistent and comparable, but it will also allow CBS to publish more figures. From now on, business confidence indices will be calculated for three additional industries, namely: Agriculture, forestry and fishing (SBI A), Culture, sports and recreation (SBI R) and other services (SBI S). In addition, the two chosen questions will be surveyed each month for most industries, resulting in monthly business confidence figures. Due to the introduction of the new method, business confidence figures for manufacturing will no longer be the same as producer confidence figures. These figures are calculated on the basis of a European definition of producer confidence, specifically for manufacturing, which allows comparisons to be made between different European countries. The producer confidence results will remain available in the existing StatLine table.

The effect of the new method

The chart below shows the business confidence indices for the non-financial private sector as a whole according to the old and the new method. Both series generally follow the same trend. The series based on the new method does, however, show a steeper decline between mid-2018 and the start of the coronavirus crisis. The biggest difference between the two series can be seen in 2022: the new index shows a continuous decline, while the old one increases in the second quarter of this year, due to a slump in respondents’ sentiment regarding the future economic climate. This is an indicator in the new method, but not in the old one. In the old method, expected sales played a major role. These rose sharply in the second quarter of 2022. The discrepancies between the two indices are due both to the methodological differences and the three additional industries that have been added under the new method. However, because these additional industries are relatively small, their effect on the overall figure is limited.

Business confidence
JaarMaandNew methodOld method
2012January -26.3-6.3

Amongst the individual industries, the differences between the old and new indices are larger for some than for others. For instance, business confidence for the construction industry is significantly lower in early 2023 under the new method, while this difference is less pronounced for the service industries. This can partly be attributed to the fact that the composition of the new indicator is more similar to the old one for the service industries than for construction, for example.

Use of the Netherlands Business Survey

The Business Survey itself has not changed. In StatLine, the outcomes for each individual question are broken down into three tables: branch of industry, region and size class. The new method for measuring business confidence will be applied from July 2023. It has also been applied retrospectively, going back to January 2012. The new tables for business confidence by branch of industry and region are available in StatLine, and the table showing business confidence by size class can be viewed in MKB-StatLine. The business confidence tables calculated using the old method will be archived, and the historical series will remain available. It should also be noted that the EIB will continue to use the old calculation method for its own sentiment indicator for the construction industry. The outcomes of these calculations will be published on